Yewon Kim
@haiyewon
Incoming CS PhD @CarnegieMellon | Music AI, Human-AI Interaction 🎹
What role can AI play in unlocking musical creativity? At #CHI2025 (🥇 Best Paper Award 🥇), we present Amuse: a songwriting tool for musicians that turns photos, sounds, and stories into chords 🎶 📄 arxiv.org/abs/2412.18940 🎧 yewon-kim.com/amuse/
Excited to announce 🎵Magenta RealTime, the first open weights music generation model capable of real-time audio generation with real-time control. 👋 **Try Magenta RT on Colab TPUs**: colab.research.google.com/github/magenta… 👀 Blog post: g.co/magenta/rt 🧵 below
So excited that Lyria RealTime is now available as an API! Try it out and fork our examples here: aistudio.google.com/app/apps/bundl…
Accepted to #ICML2025! My first physical conference in a while... Excited to see you all in Vancouver!
What do developers 𝘳𝘦𝘢𝘭𝘭𝘺 think of AI coding assistants? In October, we launched @CopilotArena to collect user preferences on real dev workflows. After months of live service, we’re here to share our findings in our recent preprint. Here's what we have learned /🧵
I’ll present our #CHI2025 paper, “Private Yet Social: How LLM Chatbots Support and Challenge Eating Disorder Recovery” in LLM for Health session. 🕤9:36 AM Wed 30. Apr 📍Annex Hall F206 📃dl.acm.org/doi/pdf/10.114… Grateful to Taehan, @subin_hci, Prof. Jennifer G Kim, and @wewantsj
A lot of big announcements today, but one nice one is the real-time music model many Magenta folks have been working on now has a name and a landing page. Introducing Lyria RealTime, the live interactive member of the Lyria family of models: deepmind.google/technologies/l…
Excited to share our #CHI2025 LBW paper, ""I know my data doesn’t leave my phone, but still feel like being wiretapped": Understanding (Mis)Perceptions of On-Device AI Vishing Detection Apps" in Yokohama!! 😆😆 🔗 Project link: nmsl.kaist.ac.kr/projects/vishi…
2025 has been unfolding in the best possible ways for me: 💖 I’ll be joining @CarnegieMellon as a PhD student (working with @chrisdonahuey!) this fall, 💖 my first first-authored #CHI2025 paper got accepted, 💖 and it received a Best Paper Award! More on the paper coming soon🎵
Thrilled to share that my *incoming* PhD student @haiyewon 's work on multimodal inspiration in music AI has been recognized with a Best Paper Award at #CHI2025 🎉 Yewon really knocked it out of the park here. Can't wait to see what she does for her PhD! arxiv.org/abs/2412.18940
🎵💿Built an MCP that lets Claude talk directly to Ableton. Now you can create music with just prompts! Here’s a demo of me creating a lush, 80s synthwave track in just two prompts. It picks the right instruments, creates melodies, and adds effects like reverb and distortion 🔊
📢 Looking for current research on #HCI + #AI? Here's a collection of 200+ #CHI2025 preprints, collected via arXiv and your suggestions: medium.com/human-centered…
What do developers 𝘳𝘦𝘢𝘭𝘭𝘺 think of AI coding assistants? In October, we launched @CopilotArena to collect user preferences on real dev workflows. After months of live service, we’re here to share our findings in our recent preprint. Here's what we have learned /🧵
Introducing Copilot Arena - Interactive coding evaluation in the wild. Our extension lets you test top models for free, right in VSCode. Let's vote and build the Copilot leaderboard! Download here: marketplace.visualstudio.com/items?itemName… Led by @iamwaynechi and @valeriechen_ at CMU. 1/🧵
Mikey Shulman (@MikeyShulman) is the co-founder and CEO of @SunoMusic, which is the world's leading AI music company and has raised more than $125M. It's also one the most beloved and fastest-growing AI consumer products of all time. Here's the full video of my most recent…
Current AI-assisted writing interfaces are disruptive. They are either too rigid, too complex, or chat-based, forcing writers into an unnatural back-and-forth. @mominnsiddiqui developed a new layered interface paradigm that makes AI-assisted writing more intuitive #CHI2025 [1/n]
Generative Text to MIDI AI for Ableton Beta is out now, give it a try!
My PhD Thesis "Improving Controllability and Editability for Pretrained Text-to-Music Generation Models" is released! ✨The thesis also includes a literature review of current control methods for text-to-music models (by July 2024). arxiv.org/abs/2411.12641
Spotify quietly launched text-to-playlist in the U.S. ✨ You give it a prompt - it can be quite lengthy! - and it generates a playlist for you. It can also incorporate your listening data (e.g. "feature songs from my recent favorite artists")
Human Creativity in the Age of LLMs Randomized Experiments on Divergent and Convergent Thinking "Our findings reveal that while LLM assistance can provide short-term boosts in creativity during assisted tasks, it may inadvertently hinder independent creative performance when…
Make music 🎻🎶 with art 🖼️🎨! Try out the awesome new tool released by @googlearts that let's you use paintings from the National Gallery to make evolving compositions in time. You can mix paintings like a DJ, or compose them like a score. goo.gle/NGMixtape Powered by…